1 |
>>>>> On Tue, 31 Dec 2019, David Seifert wrote: |
2 |
|
3 |
> On Tue, 2019-12-31 at 05:34 +0100, Michał Górny wrote: |
4 |
>> --- a/eclass/ruby-ng.eclass |
5 |
>> +++ b/eclass/ruby-ng.eclass |
6 |
>> @@ -137,7 +137,7 @@ ruby_samelib() { |
7 |
>> local res= |
8 |
>> for _ruby_implementation in $(_ruby_get_all_impls); do |
9 |
>> has -${_ruby_implementation} $@ || \ |
10 |
>> - res="${res}ruby_targets_${_ruby_implementation} |
11 |
>> ?," |
12 |
>> + res="${res}ruby_targets_${_ruby_implementation} |
13 |
>> ?(-)," |
14 |
|
15 |
> https://dev.gentoo.org/~ulm/pms/head/pms.html#x1-830008.3.4 |
16 |
> In a 4-style use dependency, the flag name may *immediately* be |
17 |
> followed by a default specified by either (+) or (-) |
18 |
|
19 |
> https://github.com/gentoo-mirror/gentoo/blob/stable/metadata/md5-cache/dev-libs/boost-1.72.0 |
20 |
> >=dev-python/numpy-1.17[python_targets_python3_6(-)?,...] |
21 |
|
22 |
> Given that you ran this through CI, this smells of dead code... |
23 |
|
24 |
AFAICS, ruby_samelib() was called only by _ruby_atoms_samelib(), and |
25 |
that function has been removed in 2010 [1]. No direct calls (ever) |
26 |
from any ebuild in the Gentoo repo. |
27 |
|
28 |
So yes, ruby_samelib() seems to be dead code. |
29 |
|
30 |
Ulrich |
31 |
|
32 |
[1] https://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo-x86/eclass/ruby-ng.eclass?r1=1.16&r2=1.17 |